1
A review on sludge dewatering indices
Vu Hien Phuong To, Tien Vinh Nguyen, Saravanamuth Vigneswaran* and Huu Hao Ngo
Faculty of Engineering and IT, University of Technology, Sydney (UTS), PO Box 123,
Broadway, NSW 2007 Australia
* Corresponding author: Tel. +61-2-9514-2641; Fax. +61-2-9514-2633; E-mail:
Abstract Dewatering of sludge from sewage treatment plants is proving to be a significant
challenge due to the large amounts of residual sludges generated annually. In recent years,
research and development have focused on improving the dewatering process in order to
reduce subsequent costs of sludge management and transport. To achieve this goal, it is
necessary to establish reliable indices that reflect the efficiency of sludge dewatering.
However, the evaluation of sludge dewaterability is not easy task due to the highly complex
nature of sewage sludge and variations in solid–liquid separation methods. Most traditional
dewatering indices fail to predict the maximum cake solids content achievable during full–
scale dewatering. This paper reviews the difficulties in assessing sludge dewatering
performance and the main techniques used to evaluate dewatering performance are compared
and discussed in detail. Finally, the paper suggests a new dewatering index, namely ―Modified
Centrifugal Index (MCI)‖, which is demonstrated to be an appropriate indicator for estimating
the final cake solids content as well as simulating the prototype dewatering process.
Key words Dewatering index; dewaterability; filterability; modified centrifugal index; sludge
dewatering
Introduction
Sewage sludge is an inevitable by–product of wastewater treatment. However, in recent
decades, the quantities of sludge produced are dramatically increasing due to the rapid growth
of industrialization and population as well as more stringent wastewater treatment standards.
According to one 2013 survey, Australia‘s total amount of sewage sludge generated was
approximately 1.3 million wet tonnes of biosolids – 200,000 tonnes greater than that in 2010
(Australian Water Association). This results in raised costs of sludge handling and transports,
which often constitutes half of treatment costs of wastewater treatment plants. Consequently,
various strategies have been made to create the best economic and environmental solutions to
this problem.
In sludge management systems, after being pre–treated by thickening, digestion or
conditioning, sludge is often dewatered before any further processing occurs such as
incineration, composting and landfill (Water Pollution Control Federation 1985). This will
result in the reducing sludge volume and, consequently, the cost of transportation (Feng et al.
2009). Despite these efforts, however, sludge dewatering is still a major challenge particularly
in designing sludge treatment systems due to the highly complex nature of sewage sludge.
Many factors influence dewatering characteristics of sludge (Karr & Keinath 1978), and this
makes it even more complicated and difficult in assessing dewatering performance correctly.
Many attempts to identify a proper indicator to truly reflect the efficiency of dewatering
2
process have been carried out by establishing relationships between these influencing factors
and sludge dewatering properties.
The terms ‗filterability‘ and ‗dewaterability‘ have been used widely to describe the
ability to dewater sludge (Sanin et al. 2011) but there is no clear distinction when using these
two parameters. Sludge filterability and dewaterability are often cited together and can be
interchangeable, which possibly results in confusion and misunderstanding. Bürger et al.
(2001) defined filtration as a mechanical method which is commonly applied for solid–liquid
separation while Mowla et al. (2013) stated that improving sludge cake filterability is one of
several ways to enhance bio–sludge dewaterability. This implies that filterability should have
been only used for measuring the of filtration process‘sefficiency instead of the whole
dewatering effectiveness. Whilst dewaterability indicates the final water content or the
maximum solid content achievable for sludge cakes, reducing sludge volume is the ultimate
objective of dewatering. Chen et al. (1996) also stated that sludge dewaterability can be
characterized by the residue moisture content in the sludge cake and the ease of the filtration
process. Yet in many published studies, sludge filterability has been considered as influencing
the dewatering process output (Scholz 2005; Yukseler et al. 2007; Sawalha & Scholz 2010).
Traditionally used dewatering indices have been mainly developed for assessing filterability of
sludge (Vesilind 2000). This may cause significant errors and inaccuracies in evaluating the
efficiency of dewatering. Consequently, besides the simplicity of the filtration process, cake
solids concentration should also be considered when assessing dewatering performance. In
other words, dewaterability indicators should have the ability to predict maximum cake solids
content achievable during full–scale dewatering.
Many studies have suggested that dewatering by filtration equipment consists of two
main phases these being filtration and consolidation or expression (Lee & Wang 2000).
Numerous investigators have put efforts on characterizing the two phases to fully describe the
full dewatering process. Sørensen et al. (1996) described sludge as a solid–liquid mixture
where solids are either in free suspension or packed closely together to form a cake. The former
is called filtration phase while the latter is known as expression phase. Novak et al. (1999)
stated that the amount of water extracted by filtration alone is not enough to achieve a desired
cake solids content; and expression phase accounts for most water removal. For compressible
structures, such as biological sludge, the filtration stage can be very short(a few minutes)
whereas the consolidation stage can be very long (several hours or even several days) to
accomplish (Saveyn et al. 2005).It has been highlighted that the high pressure in dewatering
devices has little influence on filtration phase but can help improve the water removal during
compression phase (Reichmann & Tomas 2001).Despite the fact that the importance of
expression was recognized many years ago, it has generally been neglected. This could be due
to the difficulties of separating the filtration and the expression phases during dewatering.
Olivier et al. (2007) proposed an illustration of one dimensional filtration and expression
processes to well –describe the difference between the two phases. The filtration rate of sludge
is normally described by the Specific Resistance to Filtration (SRF) which utilizes the plot of
the inverse flux (t/V) versus filtrate volume (V) to calculate it (the linear part in the filtration-
compression curve). It could be noticed that once filtration is completed and expression begins,
t/V will increase dramatically with small volume of filtrate extracted, and, to some extent, this
change can be used to differentiate the two phases. Another method to define the two stages
was proposed by Konnur et al. (2008). They stated that the expression starts at a point where
the particle network can no longer resist elastically. This results in the breakage of inter-
particle bonds and particle rearrangement, leading to irreversible consolidation.
3
Basically, sludge mechanical dewatering can be accomplished by filtration (vacuum filters,
filter presses and belt filters) and centrifugation (centrifuges) processes. Sludge dewaterability
in these two processes are essentially different (Spinosa & Mininni 1984). In fact the
techniques used to evaluate the sludge dewaterability are specific to each dewatering processes.
In this review, various methods for measuring dewatering performance that have been studied
and applied over the years are discussed and compared in terms of procedures, advantages and
limitations as well as their applications.
Challenges in measuring sludge dewatering performance
Together with improvements in sludge dewatering to achieve the highest solid content sludge
cake, it is essential to establish a reliable dewatering index that can fully express how easily
sludge releases its water (Pan et al. 2003). To date, however, there is no universal indicator yet
properly reflecting the solid–liquid separation ability of sludge. The main reasons for this
problem are linked to different aspects of sludge properties, conditioning and dewatering
methods.
Initially, as mentioned earlier, sewage sludge can vary enormously in terms of physical,
chemical and biological characteristics, leading to its relatively unpredictable behaviour,
especially dewatering behaviour (Colin et al. 1988). This makes it difficult to quantify most of
the parameters (Sanin et al. 2011). Although some parameters can be quantified, it has never
been easy to correlate these properties with sludge dewatering. Despite this impediment,
various typical sludge properties such as pH, particle surface charge, organic content, cake
porosity, compressibility, particle size, rheological characteristics, bound water content and
solid concentration – variables that can influence dewaterability of sludge – have been
investigated by numerous studies. These are summarized in more detail by Karr and Keinath
(1978). However, a consensus is still lacking on which ones have the greatest impact on sludge
dewatering.
It has been known that conditioning treatment is necessary for most sludge types,
especially for bio–sludge which has proven to be naturally difficult to dewater. Of the various
conditioning methods, polymer conditioning has been the most popular method and is often
employed in mechanical dewatering. However, many investigators have focused mostly on
optimizing the conditioning regimes but neglected the effect of dewatering equipment on
conditioned sludge (Vaxelaire & Olivier 2006). Novak et al. (1999) reported that the two
fundamental devices used for sludge dewatering, belt filter press and centrifuge, have different
conditioning demands. This implies that the selection of conditioning agents is based not only
on the nature of the bio–solids but also on the type of mechanical dewatering system.
Most traditional dewaterability measuring techniques, including Capillary Suction Time
(CST) and Specific Resistance to Filtration (SRF), often measure the rate of filtration only and
overlook the considerable contribution of expression. Therefore, to better understand the limits
of dewatering, it is necessary to focus not only on the rate of filtration but also on the rate of
expression, which is considered to be most important for achieving drier solid cakes.
The last but not least major problem with most traditional dewatering index procedures is
that they barely resemble the actual sludge dewatering processes (Lynch & Novak 1991),
except for SRF measurement, which is quite similar to pressure filters and vacuum filters
(Christensen & Dick 1987; Vaxelaire & Olivier 2006). It is known that different dewatering
devices have different operations and intensity, which greatly affects the efficiency of sludge
4
dewatering. Consequently, there should be a method that can estimate the final cake
concentration and simulate real dewatering processes. This also suggests that it is not feasible
to use only one dewaterability indicator for all dewatering processes. This is because the
liquid–solid separation is influenced by numerous parameters, and a single index is hardly
sufficient to fully describe the whole process (Vaxelaire & Cézac 2004). Hence, along with
developing accurate measurements of sludge dewatering performance, selecting a proper index
for different methods of dewatering is also significantly important (Pan et al. 2003). However,
the problem is that it is not easy to simulate the real processes occurring in dewatering
equipment. In the present work, besides SRF which can mimic filtration processes, another tool
that can be applied to other popularly used dewatering equipment, the centrifuge, is also
recommended.
Dewaterability indicators for filtration processes
Specific Resistance to Filtration (SRF)
Definition and Procedure
The SRF test was introduced in 1956 by Coackley and Jones (1956) and it became the very
first widely used technique. It is considered as an important parameter for assessing filtration
quality of sludge (Lai & Liu 2004; Qi et al. 2011). The SRF test measures the resistance of
sludge to the withdrawal of water through a porous media either by vacuum or pressure
(Graham 1999). The higher the specific resistance, the more difficult it is to dewater sludge and
vice versa (Karr & Keinath 1978). In general, sludge with SRF value as low as 1010
– 1011
m/kg
is classified as easy to dewater; whereas, sludge with SRF value as high as 1014
– 1015
m/kg is
considered as difficult to extract water. Specific resistance varies with applied pressure, filter
area and pore size and liquid viscosity, making it more complicated to measure and compare.
Figure 1 Effect of polymer dose on specific resistance to filtration (figure reprinted from Sanin
et al. (2011), © DEStech Publications, Inc.)
The SRF parameter has been utilized mainly as a measurement of dewaterability and to
optimize sludge conditioning (Figure 1). It is used for various mechanical dewatering devices
such as vacuum filters, belt presses and plate and frame presses (Christensen 1983). SRF test
was developed as a filtration model for vacuum filtration of sludge. The original measure of
SRF uses Buchner funnel apparatus. The instrument set up is shown in Figure 2a. The test is
5
conducted by pouring a reasonable volume of sludge into the funnel (with filter paper) and
applying the vacuum (measured with a vacuum gauge) at time zero. During the filtration,
filtrate volume is recorded as a function of time (Coackley & Jones 1956). These data are then
plotted with inverse flux (time/filtrate volume) versus filtrate volume (Figure 2b) and the line‘s
slope serves to calculate the specific resistance. Most of investigators determine SRF using the
following equation (Christensen & Dick 1985):
(1)
where ΔP = pressure difference(for filtration process, pressure difference refers to liquid
pressure), A = filtration area, b = the line‘s slope, µ = viscosity, = weight of dry cake solids
per unit volume of filtrate
(a)
(b) Figure 2 (a) A Buchner funnel apparatus set – up and (b) plot of the inverse flux versus filtrate
volume (figure reprinted from Sanin et al. (2011), © DEStech Publications, Inc.)
As defined earlier, the full dewatering process fundamentally consists of filtration and
expression stages. Since SRF is calculated based on the linear part of the filtration-compression
curve, it describes only filtration phase. As a result, SRF alone is not sufficient enough to
characterize the full process of dewatering (Marinetti et al. 2009).
It has been known that the conventional model of SRF test was genuinely developed for
incompressible cakes where filtration is the predominant process and SRF is a constant. Whilst,
for compressible cake where compression also plays an important role in dewatering process,
6
SRF is proved to be an increasing function of the solid pressure or compressive pressure
(Sørensen & Sørensen 1997).Generally speaking, SRF and solid pressure are two key
parameters describing the liquid flow through the cake (filtration phase) and the expression
phase, respectively. Developed from the conventional modelling, several models have been
proposed to better describe both processes of dewatering, which can be found in a review of
Lee and Wang (2000).These proposed models are based on the assumption that the cake
permeability decreases (leading to SRF increasing) as the solid pressure increases (Marinetti et
al. 2009).
Shortcomings of method
Although, SRF has been widely accepted as a useful general measurement of sludge
dewaterability, certain limitations and confusion have arisen in the determination of SRF
(Vesilind 1988). Firstly, the time required for measurements was reached when a cake was
formed and eventually cracked with a resultant drop in pressure. However, most sludge
contains fibres which prevent the cake from cracking. As a result, investigators in the 1970s
measured sludge filterability by examining the time required to obtain a given amount of
filtrate from a given initial amount of sludge (Tenney et al. 1970; Notebaert et al. 1975).
Moreover, this measurement time also depends on a number of variables such as sample
volumes, applied pressure, filter area, pore size and initial solid content of sludge, leading to
difficulties in directly comparing results obtained from different studies.
Another problem that compromises the reliability of SRF was the unclear and
inconsistent use of the parameter‘s unit. Gale (1967) reported the units commonly used by
early investigators of SRF which were the square seconds per gram (s2/g), centimetres per
gram (cm/g) and meters per kilogram (m/kg) or tetrameters per kilogram (Tm/kg). He pointed
out that s2/g was incorrect and cm/g was appropriate. However, Tebbutt (1970) latter suggested
that m/kg was a better choice for specific resistance in the SI system. Tm/kg was subsequently
recommended to eliminate the use of cumbersome scientific notation (Christensen 1983).
Most importantly, the testing method for SRF measure has been recognized as a time–
consuming and complicated procedure. It requires much expertise and sophisticated equipment
to make it function well, especially for field measurements (Barber et al. 1997; Vesilind 2000).
Capillary Suction Time (CST)
Definitions and procedures
The theory of CST was first developed by Baskerville and Gale (1968) as a substitute for SRF
test and has been increasingly popular for various applications and disciplines. CST is
principally the time required for a certain volume of filtrate drawn out of the sludge and sucked
into the blotter paper by capillary force (Vesilind 1988).. Briefly, CST can be deemed as ―time
to dewater‖ and the readings are in seconds (Novak 2006). CST has been widely used for
assessing the effects of conditioning on sludge filterability as well as determining the optimal
dose of conditioners for dewatering processes (Graham 1999). A short CST, which is less than
20s, is indicative of a readily dewaterable sludge, while a long CST is representative of a
poorly dewatering sludge. CST of autothermal thermophilic aerobic digested (ATAD) sludge
was found to exceed 50,000s (Agarwal et al. 2005).
7
As the first model regarding CST apparatus was devised in 1968 (Baskerville & Gale 1968),
various modifications of the CST test have been implemented in subsequent years. However,
the standard CST instrument basically consists of 2 plastic blocks, a stainless steel cylindrical
funnel, a Whatman No. 17 filter paper (which is a standard grade of chromatography paper), 3
electrodes fixed in the upper block and connected to an electrical timer (Vesilind 1988). The
equipment‘s installation is shown in Figure 3. The test is carried out by pouring a small amount
of sludge into the cylindrical tube. Then under the effect of capillary pressure, filtrate from the
sludge flows radially through the filter paper until it reaches the first 2 sensors that activate the
timer. The timer stops when the flow reaches the third sensor, giving the value of CST in
seconds. The capillary suction pressure is much greater than the hydrostatic pressure inside the
funnel and thus the test does not depend on the amount of sludge, as long as there is a sufficient
quantity provided to perform the test (Scholz 2005).
Figure 3 The standard apparatus (Model 304B CST, Triton Electronics Ltd.)
As can be noticed, CST is also a function of various parameters such as filter paper properties,
instrumental properties and sludge–related properties (Vesilind 1988) in the following form:
[
] (2)
where = a dimensionless instrument constant, C = solid concentration, χ = filterability
constant
Shortcomings of method
Unlike SRF, CST is affected by the concentration of solids, and therefore it is not meaningful
to compare CST of different sludge types from different plants (Vesilind 1988). As a response,
a normalized value of CST was proposed and applied for the purpose of comparison. The value
is determined by dividing CST value by the initial total suspended solids (TSS) concentration
of sludge and expressed in unit of seconds per litre per gram TSS (Yu et al. 2008).
Besides, although the CST method has been effective in characterizing a majority of
sludge, it is not often used in determining optimum polymer dose (OPD) for well–flocculated
sludge which occurs around OPD point. This is attributed to the too–fast escape of water from
the floc (Scholz 2005; Sanin et al. 2011). For this reason, Vesilind (2000) developed a new
8
CST measurement using a modified CST apparatus that allows the filterability of super–
flocculated sludge to be evaluated. Here the procedure is carried out by first draining the free
water from the well–conditioned sludge and then leaving only flocculated sludge exposed to
the filter paper.
Another emerging problem concerning CST measurement is related to the use of
standard filter paper, specifically the Whatman No. 17 chromatographic paper. Although this
paper has been used worldwide, some disadvantages were noted including anisotropic
properties, relatively over–sized pores and economic concerns. Conversely, less expensive
filters such as the Fisher 200 chromatographic papers, SS1107 and SS3205, chromatographic
papers can be used without significantly changing the range of expected CST values (Sawalha
& Scholz 2007).
Despite various efforts to improve the CST test‘s precision and reliability, the method
has still encountered many unresolved problems. For instance, it was reported that using CST
and SRF tests to determine OPD could lead to over–dosing, especially in the case of polymer
conditioning. Christensen et al. (1993) explained that it is because the calculations of both
modified CST and SRF used the filtrate viscosity which is directly proportional to the
conditioner dose, causing the measures to be inaccuracies. Another major reason for the
unreliability of CST results is that since little pressure is a on the sludge floc during
measurement, CST cannot truly reflect the floc strength or resistivity to shear during
dewatering stage (Pan et al. 2003).
Relationship between CST and SRF
As mentioned earlier, CST was developed as a substitute for SRF to measure dewatering rate.
However, since CST is only an empirical method, it is unable to predict the cake solids content
as well as simulate the real dewatering process compared to SRF. Vesilind (1988) discussed
that failures in relating CST to SRF were due to the fact that SRF should correlate with
filterability constant (χ) instead of CST. He also stated that SRF and χ are both fundamental
measures of dewaterability.
Despite of that, various studies have tried to investigate and model the relationship
between CST and SRF with the purpose of obtaining the averaged specific resistance of
filtration cake from the data generated by CST tests. Lee and Hsu (1994) proposed a method
that allowed SRF to be calculated without the liquid invasion volume measurement using
capillary suction apparatus. A similar work by Herwijn et al. (1995) presented a newly
developed model of CST apparatus able to determine specific cake resistances of both
unflocculated and flocculated sludges. Sawalha and Scholz (2010) provided a mathematical
model which related CST, SRF and other parameters, such as temperature and solids content.
These can also predict the results of SRF tests from those of CST tests. Most recently, Peng et
al. (2011) obtained a relatively good correlation between normalized CST and SRF (R2
=
0.9450). They concluded that it is not necessary to use both these parameters simultaneously to
evaluate the dewatering rate.
Methods for measuring the extended phase of dewatering using filtration processes
Novak et al. (1999) emphasized that the change in cake volume and liquid pressure are more
appropriately indicative of expression phase. Figure 4 shows that expression occurs as the cake
depth begins to decrease and the liquid pressure is no longer constant. By separating the
filtration from the expression, they also recommended an energy–saving solution for
9
dewatering equipment, especially for belt press, so that filtration stage could be executed under
very low pressure such as gravity.
Figure 4 Two phases of sludge dewatering (Novak et al. (1999), with permission from ASCE)
This research group developed a method of measuring the rate of expression using
modifications of SRF test, as described by Sørensen and Sørensen (1997). The procedure
consisted of 2 steps – gravity drainage and expression. The normalized liquid pressure, defined
as the ratio of the liquid to total pressure (pl/ptotal), was measured with time until it reached
plateau in the expression phase. Effects of applied pressure, cake thickness and compressibility
and conditioner choice on expression were also investigated. The study results highlighted that
pl/ptotal was an excellent indicator for the extended phase of dewatering. In addition, a
relationship of pl/ptotal with cake solid concentration was also established for both conditioned
and unconditioned sludge, which revealed that the ultimate dry solids content was the same for
all cases. However, the major difference was the amount of time in which the maximum solids
cake could be achieved in each case, which means the difference in the rate of dewatering
process.
Similar to the work conducted by Novak‘s group, Tastu (2007) presented a two–step
method (filtration and pressing steps) to characterize sludge dewatering properties as described
by Bouskova and la Cour Jansen (2006) and Bouskova et al. (2006). The dry solid content of
sludge after filtration (DSfiltr) and expression (DSpress) steps were employed to characterize
sludge filterability and compressibility, respectively.
Both lab–scale and full–scale results of the study demonstrated that DSpress seems to be a good
characterization parameter of sludge compressibility as well as good indicator of sludge
dewaterability. However, the study failed to characterize sludge filterability using DSfiltr. Thus,
it was concluded that filtration step could only considered as a way to prepare sludge for the
pressing stage in this study. Another limitation of the method was that the measures of DSfiltr
and DSpress were sensitive to the filtration time and the amount of sludge used for measurement.
10
In general, these two techniques were developed for filtration. It has been noted that the
procedures used the solid content of sludge cake to indicate dewatering efficiency, which
targeted the most important purpose of dewatering process. On the other hand, like all
traditional SRF tests, the above two methods experienced problems with setting suitable
operational parameters such as filtration time, the measured amount of sludge and applied
pressure. As a result, more studies should be done to develop a comprehensive method for
sludge dewatering characterization.
Dewaterability indicators for centrifugation processes
Apart from filtration devices, centrifuge has been used widely for sludge dewatering and has
become increasingly popular by virtue of its high performance in producing higher cake solids
content that can be up to 30% for dewatering of anaerobically digested biosolids (Higgins et al.
2006). As a consequence of differences in operation and conditioning requirements, filtration
and centrifugation effectiveness should be evaluated separately. It has been proven that
numerous factors influence centrifugability and various attempts have been made to determine
reliable indicators for sludge centrifugation performance. However, no appropriate parameter
has yet been developed due to difficulties in reproducing the processes taking place inside the
full–scale centrifuge in the laboratory context (Spinosa & Mininni 1984). Unlike filtration
types, filter skin formation of sludge during filtration which leads to the extended phase of
dewatering, may not occur in the centrifugal process. Therefore, Novak et al. (1999) suggested
that a theoretical assessment of the stresses on sludge during dewatering would be useful in the
case of centrifuge. Besides, as mentioned earlier, one of the crucial roles of dewatering process
is to increase the solid content of sludge in order to reduce sludge volume. Consequently, cake
solids content should be defined as a parameter representing the process‘s efficiency. For
centrifuge, it is generally classified that dewatered sludge having a dry solid content of
theoretically about 26 – 30%, which is commonly 20 – 25% in the industrial scale, indicates for
a good dewatering performance. On the other hand, solids concentration of 8 – 22% is
considered as bad dewaterability (Vigneswaran & Aim 1989).
Settleability, scrollability and floc strength
Regarding operations of filter presses or belt presses which apply compression for sludge
dewatering (Figure 5a), filterability is mainly used to evaluate the process performance. On the
other hand, for centrifuges, centrifugabilty which indicates the ability to dewater sludge by
centrifugation (Spinosa & Mininni 1984) could be defined as the ease of being conveyed by the
screw of the feeding sludge (Figure 5). Spinosa and Mininni (1984), as a consequence, reported
that sludge settleability, scrollability and floc strength were major sludge characteristics
influencing centrifugability. Unfortunately, no standard methods are available in which the
above properties are considered as a whole. Nonetheless, Vesilind (1970) introduced a
technique for measuring settling and scrolling properties of sludge using a lab–centrifuge and a
standard penetrometer. However, in many cases, the technique has proved to be unreliable for
activated sludge. Aveni and Lamarca (1974), who undertook comparative tests on both
laboratory and full–scale devices, discovered relatively good relationships between the two
cases in terms of percentage solids recovery.
11
(a)
(b)
Figure 5 The schematic diagrams of (a) belt press filter and (b) centrifuge (reprinted from
Wakeman (2007) with permission from Elsevier)
The method of floc strength measurement allows the sludge flocs‘ specific resistance to
centrifugation to be evaluated (Spinosa & Mininni 1984). This method was carried out by
measuring the CST of the stirred sludge at 1000 rpm for different stirring times. The study
concluded that a clear centrate can be obtained when the plot of CST versus stirring time is
linear with a slight slope, between 10 and 100 s, and the CST value at 10s stirring is around 10
– 12 s.
Compactibility
Compactibility, which was defined as cake solids content of sludge after centrifugation, has
been used by a number of studies (Erdincler & Vesilind 2000; Emir 2002; Emir & Erdincler
2006) to indicate sludge dewaterability along with conventional dewatering indices.
Compactibility was evaluated by centrifuging sludge samples. Following this the degree of
compaction was determined in terms of solids cake content of the compacted samples and the
height of the compacted sludge layer. Emir and Erdincler (2006) stated that compactibility is a
useful measure to assess dewaterability of hard–to–filter sludges. They also suggested the use
of this parameter together with CST and SRF to identify the most suitable dewatering method
for a given sludge. Nevertheless, their methods cannot quantify the stress imparted on sludge
cake during dewatering by centrifuge. Chu and Lee (2001) introduced an arm–suspended
centrifuge (Figure 5) to investigate the centrifugal separation of moisture from conditioned
activated sludge and determine an optimal rotational speed for maximum moisture removal.
However, the final cake solids achievable cannot be predicted using this method.
12
Figure 6 Schematics of the arm–suspended centrifuge (reprinted from Chu & Lee (2001) with
permission from Elsevier)
Modified centrifugal index (MCI) – A new centrifuge based laboratory scale
sludge dewatering
It was found that the stresses imparted to sludge during dewatering have a significant impact
on dewatering efficiency in terms of solids cake content. Higgins et al. (2006) utilized Gt value
in determining the effect of shear or mixing intensity on OPD using a calibrated lab–scale
mixer. Here, G was velocity gradient (s-1
) and t is time of mixing (s). By using this
dimensionless parameter, they determined the stress of full–scale dewatering devices using
shear stress as equivalent. However, there is a major drawback that the final cake solids
achievable at full–scale cannot be predicted with this bench–scale method (Dentel & Dursun
2009). Besides, the shear imparted on the sludge cake during dewatering is different from that
applied to the sludge liquid during conditioning.
Recently, a modified lab–scale centrifuge device, namely modified centrifugal index
(MCI) was suggested and investigated to overcome the difficulties encountered in the
traditional dewaterability indicators. The method proposed by Higgins and colleagues at
Bucknell University (Higgins et al. 2014) has been used by To et al. ( 2014) to evaluate the
sludge dewaterability of centrifugation. The stress of centrifuge, or the centrifugal force, is
measured using a dimensionless parameter gt, which is the product of times gravity g (which is
related to centrifuge rotating speed and rotor radius) and centrifugation time t (s).
(b)
(a)
Figure 7 Cake solids of (a) unconditioned sludge and (b) conditioned sludge after centrifuge as
a function of gt values (To et al. 2014)
y = 7.9073ln(x) - 81.273 R² = 0.8068
0
5
10
15
20
25
30
35
40
0 500000 1000000 1500000
Cak
e s
olid
s co
nte
nt
(%)
gt
y = 5.1324ln(x) - 54.449 R² = 0.7596
0
5
10
15
20
0 200000 400000 600000 800000 1000000
Cak
e s
olid
s co
nte
nt
(%)
gt
13
Figure 7 illustrates plots of cake solids content of unconditioned and conditioned sludge at
different gt values. These graphs indicate that the increase in gt results in the improvement of
cake solids, which implies that higher the centrifugation intensity, the better the dewatering
properties will be in terms of cake solids. However, beyond a certain value of gt, the
percentage of cake solids remained almost the same despite increasing intensity. Furthermore,
it was noticed that without conditioning, the maximum cake solid achievable by centrifuge was
only about 16% (Figure 7a). After conditioning, this cake solid increased to around 30%
(Figure 7b), which is quite similar to the dewatered sludge concentration at the wastewater
treatment plant studied. This demonstrates that MCI test can be effective in estimating the final
cake concentration as well as simulating the real dewatering process. In addition, one notes
from Figure 7 that the maximum solid cake achieved without conditioning (Figure 7a) can be
obtained at a much lower gt value when conditioning was employed (Figure 7b). It illustrated
the positive effect of polymer conditioning treatment prior to dewatering. This method has the
potential to determine the optimal polymer dose in sludge conditioning. However, this study
only worked on anaerobically digested sludge. Therefore, further studies for other sludge types
are required to prove this concept.
Other dewaterability measurement techniques
Moisture distribution
Sewage sludge is hydrophilic by character and typically has 98 to 99.7% moisture content
which is generally difficult to remove (Smollen 1990). Reducing sludge volume is basically
achieved by maximizing the cake solid content, or in another words, minimizing the water
content. Hence, a comprehensive understanding of water distribution in sludge and their
relationship to sludge dewaterability may be useful for improving the dewatering system‘s
performance.
Water classification and measurement techniques
The presence of solids causes the non–homogeneity of water within sludge, leading to different
types of water with different behaviours (Vesilind 1994). Several researchers proposed a
number of ways to classify water and most of them based on the interaction energy binding or
structural binding between water molecules and solid materials (Yin et al. 2004). Vaxelaire and
Cézac (2004) reviewed moisture distribution in different activated sludge and reported
comprehensively various ways of water classifications. The most popularly used categorization
was devised by Vesilind and Hsu (Vesilind & Hsu 1997), which defined four types of water in
sludge (Figure 8) as follows:
Figure 8 Schematic model of various forms of water in biosludge (reprinted from Mowla et al.
(2013) with permission from Elsevier)
14
- Free water: water not associated with solid particles and is separated easily by simple
gravitational settling.
- Interstitial water: water trapped inside floc structure or a cell and only a small amount
of this water might be removed by mechanical dewatering devices such as vacuum
filters or centrifuges.
- Surface (or vicinal) water: water physically held on to the surface of solid particles by
adsorption and adhesion and cannot be separated by any mechanical means.
- Hydration water: water chemically bound to the solids particles and can be released
only by thermo–chemical destruction of the particles at temperature above 105oC.
Another more simple classification is to divide water into two main types: firstly, free water
which is not influenced by solid particles; and secondly, bound water properties which are
modified due to the presence of solid particles. The bound water content is deemed to be one of
the major limiting factors affecting sludge dewatering efficiency because removing it requires
much larger amounts of energy. Over the year, various techniques have been proposed to
measure the moisture distribution in general and the bound water content in particular. These
have been compared and discussed in terms of operating conditions by Vaxelaire and Cézac
(2004). Among these techniques, the most commonly used are:
- Drying test: this test is based on the analysis of a drying curve with an assumption that
the rate of evaporation of the waters depends on the type of bond between water and the
solid particles.
- Dilatometry: this technique is based on freezing properties of waters, which measures
the freezable water (free water) content by using dilatometer placed at -20oC. The
difference between total water content and the free water content is considered to be
bound water.
- Expression test: this method is mainly used to measure bound water content which is
defined as the final moisture content of sludge expressed under a very high constant
pressure (generally around 31MPa).
- Centrifugal settling test: in this test, sludge is centrifuged at very high rotational speeds
of 3500 – 4000 rpm and the water content of sludge sediment is reported as bound
water content.
- Differential thermal analysis (DTA) and Differential scanning calorimetry (DSC): both
techniques assume that bound water does not freeze below the given threshold
temperature such as -20oC. DTA and DSC, respectively, measure the temperature
difference and the flow heat difference between sample and a thermally inert material
(reference material) (Erdincler & Vesilind 2003; Lee & Lee 2004; Deng et al.2011).
These two methods can be considered as fast and valuable tools for bound water
determination (Katsiris & Kouzeli-Katsiri 1987), however, Lee & Hsu (1995) pointed
out that they are not practical for measuring the water distribution of activated sludge
due to the sludge‘s non-uniformity. Furthermore, the selection of threshold temperature
is essentially important since it is used for distinguishing the free and bound water (Lee
& Lee 1995). As a result, DTA or DSC performed simultaneously with
Thermogravimetric analysis (TGA) (aslo known as TG-DTA and TG-DSC), as
proposed by Chen et al. (1997) and Ferrasse & Lecomte (2004), for biological products.
It is notable that these techniques imitate different methods of sludge dewatering such as
drying beds (drying test), freezing beds (dilatometry), filter presses (expression test) and
centrifuges (centrifugal settling test). However, different techniques with different principles
have led to difficulties in comparing results of different studies. It is also evident that the
various definitions of water types can cause some confusion.
15
Moisture content and sludge conditioning and dewatering
Using moisture content to assess sludge dewatering efficiency is an interesting and promising
approach that has attracted the attention of numerous researchers. Many have concentrated on
determining the correlation between moisture distribution and sludge conditioning and
dewatering.
In terms of conditioning, it has been demonstrated by a number of authors that adding
conditioner leads to a significant reduction of the bound water content, demonstrating the
positive effect of chemical conditioning on water distribution (Halde 1979; Carberry &
Prestowitz 1985). However, overdosed conditioning may cause an increase in bound water due
to the absorption of moisture onto the polymer particles (Chu & Lee 1999).
For dewatering, a relationship between dewatering energy requirement and sludge water
content (Figure 9 shows that only about 20% of water is easily removed, even for conditioned
sludge, but once water content is reduced to below 80%, the dewatering energy demand
dramatically increases (Lee & Hsu 1994; Chu & Lee 1999; Wang et al. 2010). This implies
that the sludge can no longer be mechanically dewatered to obtain a smaller residual water
content, which is also considered to be the limit of sludge mechanical dewatering. Several
studies that identified of the correlations between water distribution and sludge dewaterability
concluded that results actually depended on the measurement techniques used. For instance,
when using the drying test for measurement, no significant correlation was obtained with CST
and cake solid content (Smollen 1990). In the meantime, strong correlations with SRF
(Robinson & Knocke 1992) and cake solids content (Heukelekian & Weisberg 1956; Forster &
Lewin 1972) were detected when using dilatometry technique.
Figure 9 Relationship between sludge water content and dewatering energy demand (reprinted
from Mowla et al. (2013) with permission from Elsevier)
Apart from being a parameter used for assessing sludge conditioning and dewatering efficiency
prediction, the profile of moisture distribution within sludge may provide useful information
for choosing the dewatering methods of a specific sludge. For example, sludge with low
immobilised moisture content could be dewatered effectively by drying beds while sludge with
high immobilised moisture should be dewatered using high pressure equipment such as filter
presses or centrifuges (Smollen 1990).
16
Rheology
Rheology and sludge characterization
Rheology is the science that deals with the flow and deformation of fluids and solids under the
influence of stresses and it has become an increasingly important tool in characterizing waste
sludge, especially sewage sludge (Abu-Orf & Dentel 1999; Örmeci 2007). Although the first
studies on rheological behaviour of sewage sludge were conducted since in the 1930s, most of
them were motivated by the need to predict pumping requirements. More recent studies have
focused on examining the rheology in relation to sludge dewaterability with the aim to predict,
control and optimize conditioning and dewatering processes (Marinetti et al. 2010).
The rheological behaviours of a certain fluid can be schematically described by flow
curves, also known as rheograms. Sewage sludge is generally classified as non–Newtonian
fluid with shear thinning (pseudoplastic) behaviour and a thixotropic nature, leading to its
highly complicated rheological characteristics (Klinksieg et al. 2007). Several mathematical
models, which were mostly based on relationships between shear stress and shear rate, were
proposed to quantify the fluid rheology. Viscosity is the most basic rheological parameter.
Viscosity of the Newtonian system is constant for all shear rates while for sewage sludge,
which is a non–Newtonian system, it is a function of shear rate (Hiemenz 1986). Another
typical rheology parameter is yield stress, which is roughly defined as an initial or minimum
stress applied on material to induce true flow. The concept of yield stress and viscosity in
relation to sludge conditioning and dewatering have limited usefulness due to difficulties in
defining the terms involved in. Furthermore, it is not meaningful to compare values of different
studies, unless an adequate description of sludge types and conditioning methods are specified.
Besides these two, the maximum shear stress (τmax), indicative of the force needed to rupture
structure bonds between the polymer and sludge flocs, may possibly be used to determine OPD
along with yield stress (Marinetti et al. 2010).
Rheology and sludge conditioning
The operation of sludge conditioning has direct and profound effects on sludge rheology. For
example, the use of polymers, especially high molecular weight ones, for flocculating sludge
significantly impacts on the viscosity via increases in both floc volume and floc strength
(Dentel 1997). For this reason, Dick and Ewing (1967) suggested that sludge rheological
parameters could be used as a guide for evaluation and control of conditioning efficiency.
Various researchers have employed rheograms of conditioned sludge with different
polymer doses for determining the OPD (Örmeci 2007). Figure 10 displays two commonly
used types of rheograms for polymer dose assessment, shear stress–shear rate (Figure 10a) and
torque–time curves (Figure 10b). The former curves are obtained by using concentric–cylinder
rheometer, also known as coaxial rotational viscometer (Dick & Ewing 1967; Campbell &
Crescuolo 1982; Dick & Buck 1985) while the latter curves are obtained by the use of torque
rheometer which were developed as a response for problems associated with the former
(Örmeci & Abu-Orf 2004; Ormeci & Abu-Orf 2006; Örmeci 2007). Both rheograms show
similar curves exhibiting either a definite peak or a portion where the slope is approaching
zero. This peak indicates the point where elastic floc network bonds rupture and the higher the
peak, the more energy is needed to break up the flocs (Langer et al. 1994; Abu-Orf & Dentel
1999). Operational conditioning parameters which maximized the height of the peak are
selected as optimum. The difference between these two methods is that concentric–cylinder
rheometer only permits the deflocculation phase to be observed. In contrast, the torque–time
17
rheogram shows both flocculation and deflocculation phases since polymer injection into
sludge occurs during torque measurements (Örmeci 2007). Moreover, the torque rheometer has
been proved to have higher reproducibility than shear stress – shear rate rheograms, and they
can eliminate the need for taking sub–samples from conditioned sludges (Örmeci & Abu-Orf
2004).
(a)
(b)
Figure 10 (a) Shear stress – shear rate rheogram (Abu-Orf & Dentel (1999), with permission
from ASCE) and (b) Torque rheogram of unconditioned and conditioned sludge with different
polymer doses (reprinted from Örmeci (2007) with permission from Elsevier)
Several previous studies presented two methods based on torque rheology that can allow not
only the OPD and mixing conditions to be determined but also the best performing polymers to
be selected (Örmeci 2007). The first method uses unconditioned sludge and utilizes the peaks
observed immediately after polymer injection. This is useful for determining OPD. The second
method uses conditioned sludge and utilizes the whole torque rheograms, which is suitable for
comparing rheological and dewatering characteristics of conditioned sludges (Abu-Orf &
Örmeci 2005; Örmeci & Abu-Orf 2005). Apart from rheograms, Campbell and Crescuolo
(1983) suggested that instantaneous viscosity, which is derived from the torque or shear stress–
shear rate curves, could also be used to control sludge conditioning. Another promising method
using rheology properties to improve the sludge processing is monitoring the filtrate or centrate
(liquid stream) viscosity following mechanical dewatering by viscometer or rheometer
(Christensen et al. 1993; Dentel & Abu-Orf 1995).
18
Rheology and sludge dewatering
It is accepted that cake solids and suspended solids (SS) contents in centrate or filtrate are two
major parameters used to determine the effectiveness of dewatering operations. High cake
solids content and low level of solids recovery mean that the dewatering process is functioning
well (Sanin et al. 2011). However, traditional dewatering measurements only assess the former
and do not report on the latter. Rheological characteristics are indicative of floc strength;
therefore, it could be useful for evaluation of sludge dewatering efficiency with reference to
both criteria. Many researchers have been working on correlating the rheological properties,
i.e. viscosity and yield strength, with the sludge characteristics such as TS, SS, particle surfaces
and CST (Dick & Ewing 1967; Forster 1981; Spinosa et al. 1989; Christensen et al. 1993).
Klinksieg et al. (2007) reported a good correlation between full–scale dewatering cake and
rheological properties (at a shear stress of 500s-1
). Most of the rheology studies on the topic of
dewaterability were based on determining relationships between floc strength or network
strength and sludge dewatering properties (Abu-Orf & Dentel 1997; Abu-Orf & Dentel 1999;
Yen et al. 2002; Hou & Li 2003; Dentel & Dursun 2009; Marinetti et al. 2010). However, none
provided a strong correlation between rheological parameters and sludge dewaterability as
measured by standard test methods.
Several new rheological approaches are network strength measurement (Yen et al. 2002;
Örmeci & Abu-Orf 2004), dynamic oscillatory tests (Mori et al. 2006; Dentel & Dursun 2009),
creep tests (Baudez & Coussot 2001) and immobilization cell rheometry (Ayol et al. 2010).
These methods could be advanced in providing a realistic model of sludge behaviour during
conditioning and dewatering. For example, oscillatory tests with two parameters known as the
storage modulus (G‘) and loss modulus (G‘‘) help to determine the sludge behaviour is gel –
like or dispersion – like by comparing these values (Dentel 2004).
Summary
Evaluation of sludge dewaterability is critically necessary for any sludge treatment system
where optimizing the dewatering process is the aim. However, this work is relatively
challenging due to the unpredictable and elusive behaviour of all sludge types, especially bio–
sludge as well as variations in solid–liquid separation methods. Since the establishment of the
very first dewatering indices, various indicators for dewatering process have been developed
and applied over the years in tandem with dewatering technology. Despite these advances,
there is as yet no universal dewatering index which can fully reflect the ability to dewater
sludge. It is believed that a reliable dewatering index should not only simulate the real water
extraction process but also estimate the maximum solid content of sludge cake achievable.
Conventional dewatering indices seem to lack one or both of these, and consequently they
hardly express the efficiency of dewatering properly.
SRF, while it was cumbersome in equipment and proved to be time–consuming, can
nonetheless estimate the cake solid content of sludge after filtration. On the other hand, CST is
increasingly popular due to its ease of measurement; however, it fails to predict solid
concentration of dewatered cake. In fact, CST and SRF may be correlated strongly with free
water (Peng et al. 2011), which constitutes only 20% of total water content (Mowla et al.
2013). On the other hand, bound water content because it takes up the bulk of sludge total
moisture content, should also be taken into consideration. It is therefore suggested that to
properly evaluate the dewatering efficiency of sludge, different parameters should be included.
19
For example, CST and SRF, which often serve to register dewatering rate, could be combined
with bound water or cake solids content and this will indicate the extent of dewatering.
Rheology represents a potential powerful tool for controlling and optimizing
conditioning, but it is not sufficient to predict the performance of full–scale dewatering
systems. This is because the method has targeted the floc strength rather than the cake solids
content. In addition, it cannot reproduce the real dewatering process. It is also suggested that
along with developing accurate measurements of sludge dewatering performance, selecting a
proper index for different methods of dewatering is critical.
Finally, with reference to filtration equipment, SRF is possibly the most appropriate
efficiency indicator, while for centrifugation devices, compactibility and MCI are promising
dewaterability measurements. The major difference between these two techniques is that the
later can quantify the stress imparted on sludge during dewatering by centrifuge, which can
reflect the influence of dewatering equipment on how well the solid–liquid separation process
performs. However, since MCI is a new method, further studies are needed to fully assess its
application to sludge dewaterability.
Acknowledgement
First author thanks to Sydney Water Corporation and UTS International Research Scholarship
(IRS) for their financial supports for her study.
References
Abu-Orf, M. & Dentel, S. 1997 Effect of mixing on the rheological characteristics of
conditioned sludge: Full-scale studies. Water Science and Technology 36(11), 51-60.
Abu-Orf, M. & Dentel, S. 1999 Rheology as Tool for Polymer Dose Assessment and Control.
Journal of Environment Engineering 125(12), 1133-1141.
Abu-Orf, M. & Örmeci, B. 2005 Measuring Sludge Network Strength Using Rheology and
Relation to Dewaterability, Filtration, and Thickening—Laboratory and Full-Scale
Experiments. Journal of Environment Engineering 131(8), 1139-1146.
Agarwal, S., Abu-Orf, M. & Novak, J. T. 2005 Sequential polymer dosing for effective
dewatering of ATAD sludges. Water Research 39(7), 1301-1310.
Aveni, A. & Lamarca, V. 1974 Cooperative program on the characterization of the sludges:
centrifugability-activity of the Institute di Ricerca Breda. EEC COST 68.
Ayol, A., Dentel, S. & Filibeli, A. 2010 Rheological characterization of sludges during belt
filtration dewatering using an immobilization cell. Journal of Environmental
Engineering 136(9), 992-999.
Barber, J., Bullard, C. & Charles, M. 1997 Centrifuge test predicts solids dewaterability. Ind.
Wastewater 5(2), 31-36.
Baskerville, R. & Gale, R. 1968 A simple automatic instrument for determining the filterability
of sewage sludges. Water Pollution Control 67, 233-241.
Baudez, J.C. & Coussot, P. 2001 Rheology of aging, concentrated, polymeric suspensions:
application to pasty sewage sludges. Journal of Rheology (1978-present) 45(5), 1123-
1139.
Bouskova, A. & la Cour Jansen, J. 2006 Improvement of separation and dewatering of
activated sludge by using enhanced biological removal process over chemical
phosphorus precipitation. Journal of Residuals Science and Technology 3(3), 145-151.
20
Bouskova, A., Persson, E., la Cour Jansen, J. & Dohanyos, M. 2006 The effect of operational
temperature on dewatering characteristics of digested sludge. Journal of Residuals
Science and Technology 3(1), 43-49.
Campbell, H. & Crescuolo, P. 1983 Assessment of sludge conditionability using rheological
properties. In: Proceedings of an EEC Workshop on methods of Characteristics of
Sewage.
Carberry, J. B. & Prestowitz, R. A. 1985 Flocculation effects on bound water in sludges as
measured by nuclear magnetic resonance spectroscopy. Applied and environmental
microbiology 49(2), 365-369.
Chen, G.W., Hung, W.T., Chang, I.L., Lee, S.F. & Lee, D.J. 1997 Continuous classification of
moisture content in waste activated sludges. Journal of Environmental Engineering
123(3), 253-258.
Christensen, G. L. & Dick, R. I. 1985 Specific resistance measurements: nonparabolic data.
Journal of Environmental Engineering 111(3), 243-257.
Christensen, G. L. 1983 Units for specific resistance. Journal (Water Pollution Control
Federation), 417-419.
Christensen, J. R., Sørensen, P. B., Christensen, G. L. & Hansen, J. A. 1993 Mechanisms for
overdosing in sludge conditioning. Journal of Environmental Engineering 119(1), 159-
171.
Chu, C. & Lee, D. 1999 Moisture distribution in sludge: Effects of polymer conditioning.
Journal of Environmental Engineering 125(4), 340-345.
Chu, C. & Lee, D. 2001 Experimental analysis of centrifugal dewatering process of
polyelectrolyte flocculated waste activated sludge. Water Research 35(10), 2377-2384.
Deng, W., Li, X., Yan, J., Wang, F., Chi, Y. & Cen, K. 2011 Moisture distribution in sludges
based on different testing methods. Journal of Environmental Science 23(5), 875-880.
Dentel, S. & Abu-Orf, M. 1995 Laboratory and full-scale studies of liquid stream viscosity and
streaming current for characterization and monitoring of dewaterability. Water
Research 29(12), 2663-2672.
Dentel, S. & Dursun, D. 2009 Shear sensitivity of digested sludge: Comparison of methods and
application in conditioning and dewatering. Water Research 43(18), 4617-4625.
Dentel, S. 1997 Evaluation and role of rheological properties in sludge management. Water
Science and Technology 36(11), 1-8.
Dentel, S. 2004 Landmark Advances in Sludge Conditioning and Dewatering. In: 2nd IWA
Leading-Edge Conference on Water and Wastewater Treatment Technologies, p. 281.
Dick, R. I. & Buck, J. H. 1985 Measurement of activated sludge rheology. In: Environmental
Engineering, ASCE, 539-545.
Dick, R. I. & Ewing, B. B. 1967 The rheology of activated sludge. Journal (Water Pollution
Control Federation), 543-560.
Emir, E. & Erdincler, A. 2006 The role of compactibility in liquidsolid separation of
wastewater sludges. Water Science and Technology 53(7), 121-126.
Emir, E. Ü. 2002The Role of Dewaterability Measures on the Liquid-solid Separation of
Biological Sludges: Compactibility as a New Measure of Sludge Dewaterability.
Bogazici University, Istanbul, Turkey.
Erdincler, A. & Vesilind, P. 2000 Effect of sludge cell disruption on compactibility
ofbiological sludges. Water Science and Technology 42(9), 119-126.
Erdincler, A. & Vesilind, P.A. 2003 Effect of sludge water distribution on the liquid-solid
separation of a biological sludge. Journal of Environmental Science and Health 38(10),
2391-2400.
21
Ferrasse, J.H. & Lecomte, D. 2004 Simultaneous heat-flow differential calorimetry and
thermogravimetry for fast determination of sorption isotherms and heat of sorption in
environemental or food engineering. Chemical Engineering Science 59(6), 1365-1376.
Forster, C. & Lewin, D. 1972 Polymer interactions at activated sludge surfaces. Effluent and
Water treatment journal 12(10).
Forster, C. 1981 Preliminary studies on the relationship between sewage sludge viscosities and
the nature of the surfaces of the component particles. Biotechnology Letters 3(12), 707-
712.
Gale, R. 1967 Filtration theory with special reference to sewage sludges. Water pollution
control 66(6), 622-632.
Graham, T. M. 1999Predicting the performance of belt filter presses using the Crown Press for
laboratory simulation, DTIC Document, College of Engineering and Science, Clemson
University, South Carolina.
Halde, R. 1979 Sewage sludge characterization by vacuum drying. Filtration and Separation
16(3), 238-242.
Herwijn, A. J., Heij, E. J. L., IJzermans, J. J., Coumans, W. J. & Kerkhof, P. J. 1995
Determination of specific cake resistance with a new capillary suction time apparatus.
Industrial and Engineering Chemistry Research 34(4), 1310-1319.
Heukelekian, H. & Weisberg, E. 1956 Bound water and activated sludge bulking. Sewage and
Industrial Wastes, 558-574.
Hiemenz, P. 1986 Principles of colloid and surface chemistry. Marcel Decker, New York.
Higgins, M. J., Bott, C. & Schauer, P. 2014 Does Bio-P Impact Dewatering after Anaerobic
Digestion? In: 2014 Water Env. Federation Annual Conference, Austin, Texas.
Higgins, M. J., Chen, Y. C. & Murthy, S. N. 2006Understanding factors affecting polymer
demand for conditioning and dewatering, Water Environment Research Foundation.
Hou, C. H. & Li, K. C. 2003 Assessment of sludge dewaterability using rheological properties.
Journal of the Chinese Institute of Engineers 26(2), 221-226.
Katsiris, N. & Kouzeli-Katsiri, A. 1987 Bound water content of biological sludges in
relationship to filtration and dewatering. Water Research 21(116), 1319-1327.
Klinksieg, K., Moshage, U. & Dichtl, N. 2007 Rheology and dewaterability of municipal
sewage sludge. In: Moving Forward Wastewater Biosolids Sustainability: Technical,
Managerial and Public Synergy, IWA, Moncton New Brunswick, Canada, p. 155.
Konnur, P. 2008Model ICT Practices in Librarianship. Indian Academic Library Association.
Langer, S. J., Klute, R. & Hahn, H. H. 1994 Mechanisms of floc suspensions-III. The elastic
flow model. Journal of Colloid and Interface Science 57(2), 10.
Lee, D. & Hsu, Y. 1994 Use of capillary suction apparatus for estimating the averaged specific
resistance of filtration cake. Journal of Chemical Technology and Biotechnology 59(1),
45-51.
Lee, D.J & Lee, S.F. 1995 Measurement of bound water content in sludge: the use of
differential scanning calorimetry (DSC). Journal of Chemical Technology and
Biotechnology 62, 359-365.
Lee, D.J. & Hsu, Y.H. 1995 Measurement of bound water in sludges: a comparative study.
Water Environmental Research 67(3), 310-317.
Lee, D.J. & Wang, C.H. 2000 Review paper: Theories of cake filtration and consolidationand
implications to sluge dewatering. Water Research 34(1), 1-20.
Marinetti, M., Dentel, S., Malpei, F. & Bonomo, L. 2010 Assessment of rheological methods
for a correlation to sludge filterability. Water Research 44(18), 5398-5406.
Marinetti, M., Malpei, F. & Bonomo, L. 2009 Relevance of expression phase in dewatering of
sludge with chamber filter presses. Journal of Environmental Engineering 135(12),
1380-1387.
22
Mori, M., Seyssiecq, I. & Roche, N. 2006 Rheological measurements of sewage sludge for
various solids concentrations and geometry. Process Biochemistry 41(7), 1656-1662.
Mowla, D., Tran, H. & Allen, D. G. 2013 A review of the properties of biosludge and its
relevance to enhanced dewatering processes. Biomass and Bioenergy 58, 365-378.
Notebaert, F., Wilms, D. & Van Haute, A. 1975 A new deduction with a larger application of
the specific resistance to filtration of sludges. Water Research 9(7), 667-673.
Novak, J. T. 2006 Dewatering of sewage sludge. Drying Technology 24(10), 1257-1262.
Novak, J. T., Agerbæk, M. L., Sørensen, B. L., Hansen & Aa, J. 1999 Conditioning, filtering,
and expressing waste activated sludge. Journal of Environmental Engineering 125(9),
816-824.
Olivier, J., Vaxelaire, J. & Vorobiev, E. 2007 Modelling of cake filtration: An overview.
Separation Science and Technology 42, 1667-1700.
Örmeci, B. & Abu-Orf, M. 2004 Estimation of optimum polymer dose using the area under a
torque-time rheogram (totalized torque). In: Proceedings of the IWA World Water
Congress, Marrakech, Morocco.
Örmeci, B. & Abu-Orf, M. 2005 Protocol for measuring comparative sludge network strength–
Relation to dewaterability. ASCE Journal of Environmental Engineering 131(1), 80-85.
Örmeci, B. 2007 Optimization of a full-scale dewatering operation based on the rheological
characteristics of wastewater sludge. Water Research 41(6), 1243-1252.
Örmeci, B.& Abu-Orf, M. 2006 Using torque rheology for the identification of the optimum
polymer dose and the best performing polymer. Journal of Residuals Science and
Technology 3(4), 227-232.
Pan, J. R., Huang, C., Cherng, M., Li, K. C. & Lin, C. F. 2003 Correlation between dewatering
index and dewatering performance of three mechanical dewatering devices. Advances
in Environmental Research7(3), 599-602.
Peng, G., Ye, F. & Li, Y. 2011 Comparative investigation of parameters for determining the
dewaterability of activated sludge. Water Environment Research 83(7), 667-671.
Reichmann, B. & Tomas, J. 2001 Expression behaviour of fine particle suspensions and
consolidated cake strength. Powder Technology 121, 182-189.
Robinson, J. & Knocke, W. R. 1992 Use of dilatometric and drying techniques for assessing
sludge dewatering characteristics. Water Environment Research, 60-68.
Sanin, F. D., Clarkson, W. W. & Vesilind, P. A. 2011Sludge engineering: The treatment and
disposal of wastewater sludges. DEStech Publications, Inc., Pennsylvania
Saveyn, H., Meersseman, S., Thas, O. & Van der Meeren, P. 2005 Influence of polyelectrolyte
characteristics on pressure-driven activated sludge dewatering. Colloids and Surfaces A
262, 40-51.
Sawalha, O. & Scholz, M. 2007 Assessment of capillary suction time (CST) test
methodologies. Environmental Technology 28(12), 1377-1386.
Sawalha, O. & Scholz, M. 2010 Modeling the relationship between capillary suction time and
specific resistance to filtration. Journal of Environmental Engineering 136(9), 983-991.
Scholz, M. 2005 Review of recent trends in capillary suction time (CST) dewaterability testing
research. Industrial and Engineering Chemistry Research 44(22), 8157-8163.
Smollen, M. 1990 Evaluation of municipal sludge drying and dewatering with respect to sludge
volume reduction. Water Science and Technology 22(12), 153-161.
Sørensen, B. L. & Sørensen, P. B. 1997 Structure compression in cake filtration. Journal of
Environmental Engineering 123(4), 345-353.
Spinosa, L. & Mininni, G. 1984 Assessment of sludge centrifugability. In: Methods of
Characterization of Sewage Sludges Casey TJ, L'Hermite P and Newman PJ (eds), D.
Reidel Publishing Company, Dublin, Eire, 16-30.
23
Spinosa, L., Santori, M. & Lotito V. 1989 Rheological characterization of sewage sludges.
Recycling von klarschlamm 2, 177.
Subramanian, S. 2005 Digestion processes and their effect on dewatering and Bound water
content of sludge. MS Thesis submitted at Virginia Tech, Virginia Polytechnic Institute
and State University, Virginia
Tastu, Y. 2007 Evaluation of sludge dewatering properties. MS Thesis, Faculty of Engineering,
Lund University, Lund
Tebbutt, T. 1970 A note on the units of specific resistance to filtration. Water Pollution Control
(GB) 69, 694.
Tenney, M. W., Echelberger, Jr W. F., Coffey, J. J. & McAloon, T. J. 1970 Chemical
conditioning of biological sludges for vacuum filtration. Journal (Water Pollution
Control Federation), R1-R20.
To, V. H. P., Nguyen, T. V., Vigneswaran, S., Nghiem, L., Murthy, S., Bustamante, H. &
Higgins, M. J. 2014 Modified centrifugal index and sludge characteristics in assessing
sludge dewatering (accepted). In: IWA International Young Water Professional
Conference, Taipei, Taiwan.
Vaxelaire, J. & Cézac, P. 2004 Moisture distribution in activated sludges: a review. Water
Research 38(9), 2215-2230.
Vesilind, P. & Hsu, C. 1997 Limits of sludge dewaterability. Water Science and Technology
36(11), 87-91.
Vesilind, P. 1970 Estimation of sludge centrifuge performance. Journal of the Sanitary
Engineering Division 96(3), 805-818.
Vesilind, P. 1988 Capillary suction time as a fundamental measure of sludge dewaterability.
Journal (Water Pollution Control Federation), 215-220.
Vesilind, P. 1994 The role of water in sludge dewatering. Water Environment Research, 4-11.
Vesilind, P. 2000 A modified capillary suction time apparatus for measuring thefilterability of
super-flocculated sludges. Water Science and Technology 42(9), 135-139.
Vigneswaran, S. & Aim, R. B. 1989Water, wastewater, and sludge filtration. CRC Press.
Wakeman, R. J. 2007 Separation technologies for sludge dewatering. Journal of Hazardous
materials 144(3), 614-619.
Wang, W., Luo, Y. & Qiao, W. 2010 Possible solutions for sludge dewatering in China.
Frontiers of Environmental Science and Engineering in China 4(1), 102-107.
Water Pollution Control Federation. 1985 WPCF Wastewater Treatment, Sludge Conditioning,
Thickening and Dewatering. WPCF, Washington D.C.
Yen,P. S., Chen, L., Chien, C., Wu, R. M. & Lee,D. 2002 Network strength and dewaterability
of flocculated activated sludge. Water Research 36(3), 539-550.
Yin, X., Han, P., Lu, X. & Wang, Y. 2004 A review on the dewaterability of bio-sludge and
ultrasound pretreatment. Ultrasonics Sonochemistry 11(6), 337-348.
Yu, G. H., He, P. J., Shao, L. M. & He, P. P. 2008 Stratification structure of sludge flocs with
implications to dewaterability. Environmental Science and Technology 42(21), 7944-
7949.